The term "Benefit Risk Ratio" in EU MDR, do I need to present benefit risk analysis as a RATIO

[Regulustrator]

Hi guys, I have a question about the term used in MDR
Every time MDR mentions the term “Risk-benefit ratio" ?? It really means a "Ratio (a NUMBER)" ? such as "BMI" body weight index ???

For example, in MDR Annex I GSPR 2,
Reduce risks as far as possible means the reduction of risks as far as possible without adversely affecting the "Risk-Benefit Ratio"

and also found in GSPR 3(d), evaluate the impact of information from the production phase and, in particular, from the PMS system, on hazards and the frequency of occurrence thereof, on estimates of their associated risks, as well as on the overall risk, "Risk-Benefit Ratio" and risk acceptability;

The word "Ratio" could cause confusion...
My client have heard from their auditor and the auditor suggest them to use the "SEMI-QUANTIFY" method to treat benefits (just like we do for each risk into the different degree of harms) and then come up a ratio based two levels (e.g. risk level/benefit level).

It that reasonable? How can we make sure a "Risk-Benefit Ratio" by 0.5 or >1 is acceptable or not? (1 still can mean nothing)

In our opinion, an objective literature review of clinical data or survey from clinical experience ( like the methods used by FDA on the benefit-risk determination guidance for 510k or PMA) can provide better objective evidence than a "ratio”

Anyone provide suggestion is always welcomes

 

[planB]

The MDR also uses terms such as "benefit-risk analysis", "benefit-risk profile" or "benefit-risk determination" in a similar context as "benefit-risk ratio". In my opinion, the "ratio" is acceptable when the benefit of the device is evaluated to outweigh the risks. There is no restriction that this evaluation/analysis cannot be performed in a descriptive, qualitative way.

 

[yodon]

Agree with @planB - and, in fact, if you look at the guidance (24971:2020), they suggest framing benefits in terms of aspects like magnitude and probability the benefits will be realized. And when doing the comparison of benefits / risks, they talk to things like uncertainty and comparisons to similar devices. So it's expected, I believe, to be more of a (perpetual) "court case" where you establish the "evidence" for the benefits, compare with the "evidence" of the risks, and can hopefully come to a logical conclusion that the benefits outweigh the risks.

There's also a good discussion of approaches in section A.7.2) in MEDDEV 2.7/1 rev 4.

 

[Marc]

MEDDEV 2.7/1 rev 4.

This link is broken.

 

[yodon]

Here's the link: https://www.medical-device-regulation.eu/wp-content/uploads/2019/05/2_7_4_en.pdf

 

[d_addams]

The short answer is no. IMO the use of the word ratio is just an artifact of claiming to weigh two things against each other so from that aspect their relative 'weights' could be expressed as a ratio conceptually, not literally.

The long answer is: One of the major issues with Risk Management is that it often uses numbers as the outputs of entirely qualitative assessments. So risk management looks and feels like a qualitative assessment such that people get wrapped around the axel thinking 'If I make the benefit estimate quantitative, I can come up with an actual quantitative ratio'. Medical risk is a qualitative assessment as well as medical benefit. 'Proving' this is to ask, do all physicians and all patients make the same decision as to the acceptability of risk for a procedure? The answer is hopefully clear to you as 'no'. Therefore, why would one think they can boil down the risk management process output to a single number that all people could agree is the 'correct number'.

If one attempted to do so, you could perhaps say that if the benefit risk is over 10 (however you arrived at that) no further review is needed (say the use of a band-aid or tongue depressor). But the closer you get to 1.0, more review/scrutiny is required. At a B/R of 1.001, it seems the benefit outweighing the risk is debatable and we should really heavily scrutinize the decision with may independent reviewers to determine if the benefit really outweighs the risk. But to say well 1.0001 is greater than 1, we're good to ship millions of product forever, is not what is intended.